Vol. 8: 179–187, 2010 AQUATIC BIOLOGY Published online March 9 doi: 10.3354/ab00233 Aquat Biol

OPENPEN ACCESSCCESS

Allometry and in the chela shape in the squat lobster rugosa

Thomas Claverie1, 2,*, I. Philip Smith1

1University Marine Biological Station Millport, Isle of Cumbrae KA28 0EG, UK 2Present address: Biology Department, 332 Morrill Science Center South, 611 North Pleasant Street, University of Massachusetts Amherst, Amherst, Massachusetts 01003, USA

ABSTRACT: Sexual selection generally promotes the development of ornaments or weapons that improve an individual’s chance of mating. The squat lobster Munida rugosa exhibits a range of vari- ation in chela (claw) morphology, with some individuals having a particular arched chela morpho- logy, apparently specialized as a weapon. Geometric morphometric techniques were used to com- pare chela morphology across a wide size range and between sexes. The most fully developed arched morphology only occurred in large males, although not all large males possessed an arched claw, while smaller males retained straight, slender claws. Some large females exhibited a reduced version of the arched morphology. The pattern of occurrence of the arched morphology suggests that it is, or has been, subject to sexual selection through male–male competition for mates.

KEY WORDS: Munida rugosa · Secondary sexual dimorphism · Chela shape · Sexual selection · Weapon · Geometric morphometric

Resale or republication not permitted without written consent of the publisher

INTRODUCTION Emlen 2008). Growth of such structures is often posi- tively allometric, with weapons or ornaments becom- Evolution of morphology in can be directed ing disproportionately developed in larger or older in- by natural selection (Darwin 1859) and sexual selec- dividuals (Petrie 1988, 1992, Simmons & Tomkins 1996, tion (Darwin 1871). In a given , pressures di- Knell & Fortey 2005, Emlen 2008), but allometry is nei- rected by natural selection favour morphologies that ther necessary nor sufficient evidence for a sexually facilitate individual survival, growth and reproduction, selected trait (Bonduriansky 2007). Evolution of allom- leading to an increase in fitness in a given environ- etry in such traits requires individual condition to be ment (the ‘struggle for life’, Darwin 1859). Pressures related to body size and the selected trait to be under associated with sexual selection directly promote traits directional selective pressure (Bonduriansky 2007). that facilitate success in mating either through intra- Emlen (2008) also noted that the morphology of all ex- sexual competition or mate choice (Kodric-Brown & aggerated male weapons vary greatly among individu- Brown 1984). For example, across a wide range of taxa, als within populations. Consequently, based on empiri- weapons in males appear to have evolved through in- cal evidence (Petrie 1988, 1992, Simmons & Tomkins trasexual selection (Emlen 2008). There are costs in de- 1996, Knell & Fortey 2005, Emlen 2008), a trait acting veloping such traits and only an increase in overall fit- as a weapon has a good chance of being a sexually di- ness can explain the evolution of a costly morphology morphic structure that is positively allometric and vari- (Bonduriansky & Day 2003, Emlen 2008). In general, able in shape or size in mature males. the magnitude of a sexually selected trait such as a In decapod , the first pereiopod generally weapon is related to the quality of the individual, in bears a chela (claw) formed by a modification of the terms of competitive ability (i.e. honest signalling, dactylus and propodus and is known as a cheliped.

*Email: [email protected] © Inter-Research 2010 · www.int-res.com 180 Aquat Biol 8: 179–187, 2010

This appendage is typically used in feeding, defence vations indicate that large males chase other males against predators and agonistic behaviour, and has away from their vicinity and they are often surrounded therefore been considered a multifunctional organ by females (T. Claverie pers. obs.). A closely related (Brown et al. 1979, Lee 1995). When this appendage is species, M. sarsi, has a similar claw morphology; claws used as a weapon, it is to be expected that dominant are used in agonistic encounters involving cheliped individuals will have larger or stronger chelipeds and grabbing (Berrill 1970) and male M. sarsi use chelipeds this has indeed been found in the few studies in which to restrain females during mating (Pothanikat 2005). this has been investigated (Barki et al. 1997, Sneddon Assuming that the arched chelae are better weapons et al. 1997, 2000). Furthermore, allometric growth and than straight chelae (Claverie & Smith 2007) and that sexual dimorphism in cheliped size is common in deca- agonistic interactions between males are primarily for pods, with males having larger chelipeds than females access to females (intrasexual selection), we predict (Hartnoll 1974). that arching is a sexually dimorphic feature with a high The squat lobster Munida rugosa has a peculiar che- degree of variation in mature males. However, it is not la morphology that can fulfil various functions, such as clear to what extent the amplitude of arching is corre- fighting and feeding (Claverie & Smith 2007). There is lated to cheliped size, nor the relative contributions of sexual dimorphism in cheliped length (Claverie & these attributes to agonistic success. Smith 2009) but, in addition, some males develop a Left–right asymmetry in chela shape (heterochely) is modified form (arched morphology, Fig. 1A) that is frequently observed in decapods (Crane 1975, Lee stronger and appears to be adapted to inflict puncture 1995, Mariappan et al. 2000) and is generally associ- wounds on other males during intense agonistic inter- ated with adaptations for capture and handling of prey actions (Claverie & Smith 2007). Moreover, field obser- (Govind & Pearce 1989, Lee 1995) but, as in the case

Fig. 1. Munida rugosa. Different chela morphologies encountered in Munida rugosa, (A) arched and (B) straight, and (C) associ- ated landmarks (1 to 6) whose coordinates were recorded from each chela for analysis of shape. Drawing modified from Claverie & Smith (2007) Claverie & Smith: Munida rugosa claw morphology 181

of fiddler crabs, can also be adapted for agonistic made jig. A small translucent elastic band was used to behaviour and female attraction (Crane 1975, Levinton hold the chela closed. Photographs were taken with a et al. 1995). Heterochely is generally triggered at an digital camera (Fujifilm Finepix F810 in a WP-FX701 early stage in development, either because of the loss waterproof housing) in macro mode and mounted on of one cheliped or differential usage between left and the jig 25 cm vertically above the cheliped to standard- right chelae (Govind & Pearce 1989, Young et al. 1994, ize the picture and minimize distortion. The prodopus Goldstein & Tlusty 2003). Moreover, in heterochelous and dactylus are only slightly curved laterally, so species, homochelous specimens are rare (if they exist analysis of chela shape in 2 dimensions was considered naturally at all) and occur generally only when animals adequate. Cheliped length of each was mea- are raised in artificial situations, such as in isolation sured from the photographs using the image analysis (Govind & Pearce 1989, Goldstein & Tlusty 2003). Con- software SigmaScan Pro 5.0.0. sequently, heterochely can be an important morpho- Variation in chela morphology. Geometric morpho- logical feature in crustaceans arising from an underly- metric methods (Zelditch et al. 2004) were used to in- ing ontogenic mechanisms linked with the functional vestigate variation in chela shape. Images of the right significance of particular chelae designs. cheliped were reflected (in Adobe Photoshop Ele- Claverie & Smith (2007) argued that the arched mor- ments 2.0) to appear like the left cheliped to minimize phology of Munida rugosa was better adapted to inflict digitising error (Rosenberg 2002) and to simplify sym- injuries while retaining functionality for feeding (prob- metry analysis (Klingenberg & McIntyre 1998, Klin- ably with the same effectiveness). Also, M. rugosa was genberg et al. 2002). ‘Landmark’ coordinates were re- observed to use both chelipeds during agonistic inter- corded (pixel coordinates were subsequently scaled to actions (T. Claverie pers. obs.), as described for distances in mm) from the image with TpsDig2 (Rohlf M. sarsi (Berrill 1970). Consequently, it is unlikely that 2005a). The 6 landmarks used were the same as those selective pressures that normally act on decapod che- used previously (Claverie & Smith 2007; Fig. 1C). They liped asymmetry (i.e. feeding ecology in durophagous were: (1) the tip of the pollex; (2) the junction between crabs or agonistic interaction in fiddler crabs) would the pollex and the manus on the inferior margin of the promote asymmetry in M. rugosa, since feeding and propodus; (3,4) the lower and upper attachment points fighting could be performed with the arched design of the carpus with the propodus; and (5,6) the upper (Claverie & Smith 2007). We therefore predict that and lower points marking the junction of the dactylus there should be no consistent chela shape asymmetry with the propodus. in this species. A principal components analysis (PCA) was performed To investigate whether patterns in chela arching in on the landmark coordinates, after procrustes superim- Munida rugosa follow our predictions and are consis- position using the TpsRelw program (Rohlf 2005b), to ex- tent with sexually selected traits, chela shape, symme- amine major trends of chela shape variation. try and allometry were analysed in relation to body Relationship between chela shape and body size. size, sex and cheliped size. Relationships between chela shape and CL, and be- tween chela shape and cheliped length, were tested with multivariate regression. Separate analyses were MATERIALS AND METHODS performed for the left and right chelae for males and females. Significance was tested using the Wilks’ Collection of specimens. Munida rugosa were col- lambda determinant ratio (Zelditch et al. 2004). lected in the Clyde Sea area, Scotland, by beam- Variability of chela morphology in relation to CL was trawling (2 m beam trawl, 50 mm mesh) on a gravel– quantified for males and females. The first principal mud sea bed at water depths ranging from 35 to 40 m component (PC1) was plotted against CL. A running during spring 2005. Individuals with both chelipeds mean of PC1 was calculated on a 3 mm CL window intact (no injury or mark of any previous moulting across the full range of CL. Squared residual variation problem) were returned alive to the laboratory for exa- (standardised to CL) of the PC1 value from the running mination. The carapace length (CL, from the base of mean was then calculated and the cumulative squared the rostral spine to the posterior margin of the carapace standardised residuals were plotted against CL. Resid- at the mid-line) of each individual was measured with uals were divided by associated CL to represent a vernier callipers to 0.1 mm accuracy. A total of 265 ani- standardized measure of the shape variation (scaled to mals were measured (130 male and 84 female), rang- the size of animal) in order to correct for the amplitude ing in CL from 13 to 38 mm. of shape difference due to animal size. A deviation A digital photograph of the lateral surface of each from a constant increase of cumulative standardised cheliped was taken. Chelipeds were held extended squared residual (greater or lesser rate of increase) and horizontal for the photograph using a custom- would indicate variation in variability of chela shape 182 Aquat Biol 8: 179–187, 2010

for a given CL. PC1 was used to describe chela shape with narrower chelae, and only males reach high val- because it explained most of the variability (Claverie & ues on the first axis, associated with high propodus, Smith 2007). long manus and short pollex. The second axis repre- Chela shape difference between sexes and appen- sents the degree of bending of the chela in the vertical dage side was tested using a 2-way permuted multi- plane, with an inflection point at the base of the pollex variate analysis of covariance (PERMANCOVA, 1000 (Fig. 2a). The third axis represents variation in chela permutations) on shape variables (partial procrustes height (Fig. 2b). distances) with sex and side as factors and cheliped length as the covariate. In order to exclude any bias from the analysis, only specimens of each sex within Relationship between chela shape and body size the same size range were included (cheliped length 54 to 114 mm). The program Manovaboard 6.4 (Sheets There was a significant relationship between chela 2006) was used for this analysis. This program uses a shape and CL for both left (Wilks’ λ = 0.35, p < 0.001) permutation test to determine significance (based on and right chelae (Wilks’ λ = 0.39, p < 0.001) in males; the sum of squared partial procrustes dis- tances — shape variables — between and within groups) rather than a conventional MANCOVA procedure (i.e. using an F-test of a variance ratio). Assessment of cheliped asymmetry. In addition to the PERMANCOVA described above, the correlations between chela shape difference (left versus right chelae) and cheliped length difference (left versus right cheliped), as well as the correlation between chela shape difference and CL, were tested with Spearman’s rank correla- tion in order to assess whether chela shape asymmetry was related to body size or che- liped length asymmetry (i.e. regeneration of one of the chelipeds, Claverie & Smith 2009). Shape differences were quantified using procrustes distance between left and right chelae (calculations were performed using the software Coordgen from the IMP series, Zelditch et al. 2004).

RESULTS

Variation in chela morphology

The first 3 principal components of the PCA together accounted for 94.7% of the shape variation (76.48% for PC1, 13.78% for PC2 and 4.44% for PC3). Examination of the thin plate spline (TPS) representations of extreme forms on each axis indicates that the first axis mainly represents an inverse relationship between pollex (distal exten- sion of the propodus) length and both manus (proximal part of the propodus) Fig. 2. Munida rugosa. Results of a principal components analysis, with thin plate spline representations of extreme landmark configurations on each length and propodus height (Fig. 2a). Data axis. Relationships between (A) principal components 2 and 1, and (B) prin- points for females are mainly clustered on cipal components 3 and 1. Filled symbols: females; open symbols: males; the negative part of the first axis, associated circle: left chela; triangle: right chela Claverie & Smith: Munida rugosa claw morphology 183

the linear model explained 35.39% of the variation of females, chela shape changed little with increasing chelae shape for both sides. For females, similar rela- cheliped length and CL (Figs. 3b & 4b). tionships were found for the left (Wilks’ λ = 0.58, p < A plot of the first axis of the PCA of chela shape 0.001) and right chelae (Wilks’ λ = 0.57, p < 0.001), but (which accounted for most of the shape variation) the linear model explained only 14.71 and 13.45% of against CL for males and females illustrates consider- the variation in chela shape, respectively. able variability in chela shape across the whole range There was also a significant linear relationship be- of sizes (Fig. 4a,b). In males, there was a pronounced tween chela shape and cheliped length for both left increase in chela shape variation at CL >30 mm (Wilks’ λ = 0.27, p < 0.001) and right chelae (Wilks’ λ = (Fig. 5a). Females, however, did not show such a dis- 0.30, p < 0.001) in males, with the linear model explain- tinct increase in chela shape variability with increasing ing 40.24 and 40.41% of the shape variation, respec- CL (Fig. 5b). tively (Fig. 3a). For females, the corresponding regres- Results from the PERMANCOVA showed that chela sions were also significant (left: Wilks’ λ = 0.60, p < shape was significantly affected by cheliped size (p < 0.001; right: Wilks’ λ = 0.59, p < 0.001; Fig. 3b), but 0.001) and differed between sexes (p < 0.001), but again the linear model explained only 12.02 and there was no interaction between cheliped size and sex 10.93% of the chela shape variation for left and right (p = 0.322), nor was there a difference between sides sides, respectively. In males, chela shape became (p = 0.437). There was still no effect of side (p = 0.230) higher with a relatively shorter pollex and a longer if all the specimens were included in the analysis manus (i.e. more arched) with increasing cheliped (including the differing size ranges of males and length and CL, on average (Figs. 3a & 4a). However, in females).

Assessment of cheliped symmetry 0.20 A Difference in shape between left and right chelae (i.e. procrustes distance 0.15 between the 2 chelae) was highly corre- lated with difference in cheliped length for

0.10 males (rS = 0.55, p < 0.001) and females (rS = 0.32, p = 0.003; Fig. 6a). However, the dif- ference in chela shape between left and 0.05 right was not correlated with CL for males

(rS = 0.15, p = 0.086) or females (rS = 0.13, p 0.00 = 0.226; Fig. 6b).

–0.05 DISCUSSION

–0.10 Variation in chela morphology in Munida rugosa has been recorded in previous stud- 0.10 ies (Ingrand 1937, Zainal 1990, Hartnoll et B al. 1992, Combes 2002, Claverie & Smith 0.05 2007), but was not investigated further than

Principal component 1 component Principal a qualitative description in relation to body size. Claverie & Smith (2007) noted varia- 0.00 tion from a straight morphology, with a nar- row propodus and straight pollex and –0.05 dactylus, to an arched morphology, with a higher propodus and bent pollex and dactylus leaving a gap in between. In the –0.10 present study, a large variety of intermedi- 40 60 80 100 120 140 160 180 ate shapes was observed from narrow Cheliped length (mm) (straight) to broad (arched) propodi, with Fig. 3. Munida rugosa. The first axis of the principal component analysis of the greatest propodus heights observed chela landmark configurations as a function of cheliped length for (A) males only in males (represented by PC1). There and (B) females. (s): left cheliped; (m): right cheliped was a positive relationship between the 184 Aquat Biol 8: 179–187, 2010

0.20 Smith 2009). It therefore appears that one A secondary sexual feature (arched chelae) 0.15 becomes prevalent 3 to 4 moults later than another (allometric cheliped growth, com- monly encountered in decapods). This 0.10 could be due to a slow development of arched morphology over several moults, 0.05 perhaps because change in chela mass takes longer than an increase in cheliped length. 0.00 Claw morphology in Munida rugosa exhibits patterns previously observed for –0.05 sexually selected traits (i.e. greater weapons in larger males, Petrie 1992, Simmons & Tomkins 1996, Knell & –0.10 Fortey 2005, Emlen 2008), assuming that arched chelae are more effective 0.10 weapons (Claverie & Smith 2007). Alter- B native chela morphologies (i.e. straight) may represent alternative reproductive 0.05 Principal component 1 phenotypes within sexes associated with alternative tactics (phenotypic outcome) 0.00 within a common strategy (decision rules, Dominey 1984). Such a strategy is com- –0.05 monly called a conditional strategy and is thought to be the most common evolution- ary stable strategy in nature (Gross 1996). –0.10 Using this strategy, males may adopt dif- 10 15 20 25 30 35 40 ferent tactics depending on their status in Carapace length (mm) the population and fitness among tactics Fig. 4. Munida rugosa. The first axis of the principal component analysis as a is not necessarily equal (Gross 1996). function of carapace length with associated running means (window of 3 mm) There is evidence for conditional stra- for (A) males (with thin plate spline representations of extreme landmark tegies in some crustaceans. In Macro- s m configurations) and (B) females. ( , solid line): left cheliped; ( , dashed brachium rosenbergii and Rhynchocinetes line): right cheliped typus, for example, dominant or alpha males are more aggressive, have well- amplitude of arching and CL and cheliped length, but developed chelipeds and are superior in competition with considerable variability. The amplitude of arching for access to food, shelter or mates. Subordinate males, depended on cheliped size (the greatest arching was on the other hand, retain relatively small chelipeds and encountered in the longest chelipeds), but the recipro- use sneaking tactics to obtain these resources (Barki et cal was not true, as the longest chelipeds were not al. 1992, Correa et al. 2003). When dominant males dis- always arched. appear, subordinate males replace them and become Chela shape varied from the straight towards the morphologically the new alpha males (Karplus 2005). arched morphology in both sexes, but the fact that the In M. rugosa, one morphology (apparently the arched degree of arching was greater in males (for equivalent one since it is exaggerated only in large males) may be female cheliped size and also because males devel- costly to develop, but could have greater benefits in oped longer chelipeds than females), suggests that terms of intrasexual competition or intersexual attrac- there is stronger selection for this character in males. In tion and could be associated with dominant males both sexes, the arched morphology appeared only in (Claverie & Smith 2007). The second morphology (in individuals larger than the morphological size of matu- this case the straight one, since it appears more like a rity (Claverie & Smith 2009). In males, however, varia- juvenile form) may be less costly to develop, but could tion in chela shape (due to development of the arched confer limited disadvantages in intrasexual competi- morphology) increased markedly above a CL of tion or intersexual attraction (e.g. not being able to 30 mm, whereas the size of morphological maturity hold firmly or injure conspecific) and would conse- indicated by cheliped length is 22 mm CL (Claverie & quently be associated with subordinate males. Costs Claverie & Smith: Munida rugosa claw morphology 185

2 1 A

) B –4 1.5 0.5

1 0 15 20 25 30 35 40 Carapace length (mm) 0.5

Fig. 5. Munida rugosa. Cumulative standardized squared Cummulative sum (× 10 residuals of the first axis of the principal component analysis 0 scores from the running mean as a function of carapace 15 20 25 30 35 40 length for (A) males and (B) females. Solid line: left cheliped; Carapace length (mm) dashed line: right cheliped associated with the arched morphology could relate to a conditional strategy). Females have shorter che- the production of larger muscular and exoskeletal lipeds than males, on average (Claverie & Smith 2009) mass and greater energetic cost to maintain a bigger and may, as a result, be less likely to develop the claw. The arched morphology could also impose a arched form exhibited by large males. However, the longer moulting time to extract the cheliped from differences in shape between males and females were exuvia, which may increase exposure to pre- dators. Intermediate morphologies may rep- 0.20 resent a development towards the arched form. Under this hypothesis, both morpholo- A gies (representing different tactics) would 0.15 likely be associated with behavioural adapta- tions appropriate to the difference in com- petitiveness (Dominey 1984). 0.10 Another possible explanation for the occur- rence of straight chelae in large males is that autotomised chelipeds may take several 0.05 moults to reacquire the arched morphology during regeneration, as suggested by Ingrand (1937) and Hartnoll et al. (1992). Intermediate 0.00 morphologies would occur during regenera- 0 10 20 30 40 tion. Studies in other decapods have shown Cheliped length difference (mm) that a particular cheliped morphology may take several moults to regenerate (Wilson 0.20

1903, Juanes & Smith 1995, Read & Govind distance Procrustes B 1997). In that case, the loss of an arched che- 0.15 liped would have a high cost, since regenera- tion would be long (this species moult onces or twice a year depending on size) and fitness 0.10 would consequently be adversely affected due to a reduction in competitive status. Similar allometric slopes were found in 0.05 the size-related change in chelae shape for males and females (i.e. non-significance of the interaction term between cheliped size 0.00 and sex). Consequently, there may be a com- 10 15 20 25 30 35 40 mon genetic basis for chela arching in both Carapace length (mm) sexes, with degree of arching being largely Fig. 6. Munida rugosa. Procrustes distance between left and right chelae determined by the length of the cheliped as a function of (A) absolute cheliped length difference between left and (and maybe social status in the hypothesis of right chelae and (B) carapace length. (s): males; (d) females 186 Aquat Biol 8: 179–187, 2010

not solely due to cheliped length, since females had Acknowledgements. This study was funded by the Sheina less arched chelae than males of equivalent size. Nev- Marshall Bequest. We acknowledge C. Klingerberg, M. ertheless, there may be intrasexual competition Zelditch and D. Swiderski for helpful discussions on geomet- ric morphometric analysis. We also thank the anonymous among females, giving rise to a certain level of selec- reviewers of this manuscript for their valuable comments. tion for this trait (Kodric-Brown & Brown 1984). If so, Thank you to the crew of the R/V ‘Aplysia’ and the divers of the limited degree of arching in females implies lower the University Marine Biological Station Millport for their selection pressure for arched chelae and correspond- help with sampling and measurements. ingly lower intensity of competition mediated through agonistic interactions using the chelipeds than in LITERATURE CITED males. No significant asymmetry was measured in chela Barki A, Karplus I, Goren M (1992) Effects of size and mor- shape of Munida rugosa and most specimens were ap- photype on dominance hierarchies and resource competi- tion in the freshwater prawn Macrobrachium rosenbergii. proximately symmetrical in cheliped length (Claverie Anim Behav 44:547–555 & Smith 2009). In the present study, some specimens Barki A, Harpaz S, Karplus I (1997) Contradictory asymme- were heterochelous but no specific handedness was tries in body and weapon size, and assessment in fighting observed, and shape asymmetry was correlated with male prawns, Macrobrachium rosenbergii. Aggress Behav 23:81–91 the asymmetry in cheliped length (with the longer Berrill M (1970) The aggressive behavior of Munida sarsi claw being the most arched). Some asymmetry in arch- (Crustacea: Galatheidae). Sarsia 43:1–11 ing was previously reported for this species (Zainal Bonduriansky R (2007) Sexual selection and allometry: a crit- 1990, Hartnoll et al. 1992, Combes 2002), but findings ical reappraisal of the evidence and ideas. Evolution 61: were not consistent. M. rugosa possess long and 838–849 Bonduriansky R, Day T (2003) The evolution of static allo- slender chelipeds and autotomise these appendages metry in sexually selected traits. Evolution 57:2450–2458 more readily when handled (particularly ovigerious Brown SC, Cassuto SR, Loos RW (1979) Biomechanics of females, T. Claverie pers. obs.) than crabs or lobsters chelipeds in some decapod crustaceans. J Zool 188: with more robust chelipeds. Hartnoll et al. (1992) 143–159 Claverie T, Smith IP (2007) Functional significance of an suggested that arching could take several moults to unusual chela dimorphism in a marine decapod: special- develop. The high prevalence of symmetric specimens ization as a weapon? Proc R Soc Lond B 274:3033–3038 in the population sampled, combined with the correla- Claverie T, Smith IP (2009) Morphological maturity and allo- tion between shape and length difference, strongly metric growth in the squat lobster Munida rugosa. J Mar suggests that the observed asymmetries in cheliped Biol Assoc UK 89:1189–1194 Combes JCH (2002) Aspects of the biology and fisheries eco- length and chela shape could be the result of regener- logy of the velvet swimming crab, Necora puber (L.), and ation after autotomy. the squat lobsters Munida rugosa (Fabricius) and M. sarsi In summary, Munida rugosa exibit a sexual dimor- Huus (Crustacea: ) in Scottish waters. PhD the- phism where males are larger than females, possess sis, University of London Correa C, Baeza JA, Hinojosa IA, Thiel M (2003) Male domi- longer chelipeds and can have a greater degree of nance hierarchy and mating tactics in the rock shrimp chela arching. However, considerable variation in the Rhynchocinetes typus (Decapoda: Caridea). J Crustac Biol degree of arching was observed for large males and 23:33–45 only some of the largest individuals fully developed Crane J (1975) Fiddler crabs of the world (Ocypodidae: arched chelae. Considering the various examples of Uca). Princeton University Press, Princeton, NJ Darwin C (1859) On the origin of species by means of natural exaggerated traits occurring in the animal kingdom selection, or the preservation of favoured races in the related to sexual selection (Zahavi 1975, Siva-Jothy struggle for life. John Murray, London 1987, Emlen & Nijhout 2000, Hongo 2003, Emlen Darwin C (1871) The descent of man, and selection in relation 2008) and similar patterns observed in other decapods to sex. Princeton University Press, Princeton, NJ Dominey WJ (1984) Alternative mating tactics and evolution- associated with male–male interactions (Crane 1975, arily stable strategies. Am Zool 24:385–396 Barki et al. 1992), it seems possible that sexual selec- Emlen DJ (2008) The evolution of animal weapons. Annu Rev tion has favoured the evolution of longer chelipeds Ecol Evol Syst 39:387–413 with a greater degree of chela arching (weapons) in Emlen DJ, Nijhout HF (2000) The development and evolution mature male M. rugosa. However, the effect of arch- of exaggerated morphologies in insects. Annu Rev Ento- mol 45:661–708 ing on the outcome of intrasexual conflict remains to Goldstein JS, Tlusty MF (2003) Substrate determinants and be quantified. It is to be expected that males with developmental rate of claw asymmetry in American lob- arched chelae would be more successful in agonistic sters, Homarus americanus. J Crustac Biol 23:890–896 interactions or in obtaining and handling mates than Govind CK, Pearce J (1989) Critical period for determining claw asymmetry in developing lobsters. J Exp Zool 249: similar-sized males with straight chelae. These pre- 31–35 dictions are readily testable with behavioural experi- Gross MR (1996) Alternative reproductive strategies and tac- ments. tics: diversity within sexes. Trends Ecol Evol 11:92–98 Claverie & Smith: Munida rugosa claw morphology 187

Hartnoll RG (1974) Variation in growth pattern between positively allometric and, if so, why? Anim Behav 43: some secondary sexual characters in crabs (Decapoda 173–175 Brachyura). Crustaceana 27:131–136 Pothanikat RME (2005) The behaviour and ecology of Munida Hartnoll RG, Rice AL, Attrill MJ (1992) Aspects of the biology rugosa and Munida sarsi. PhD thesis, Queen’s University of the galatheid genus Munida (Crustacea, Decapoda) Belfast from the Porcupine Seabight, Northeast Atlantic. Sarsia Read AT, Govind CK (1997) Regeneration and sex-biased 76:231–246 transformation of the sexually dimorphic pincer claw in Hongo Y (2003) Appraising behaviour during male–male adult snapping shrimps. J Exp Zool 279:356–366 interaction in the Japanese horned beetle Trypoxylus Rohlf FJ (2005a) TpsDig, digitize landmarks and outlines, dichotomus septentrionalis (Kono). Behaviour 140:501–517 v. 2.05. Department of Ecology and Evolution, State Uni- Ingrand M (1937) Morphologie des pinces et caractères sex- versity of New York at Stony Brook, Stony Brook, NY uels secondaires de Munida bamffica. Trav St Biol Roscoff Rohlf FJ (2005b) TpsRelw, relative warps, v. 2.05. Department 15:57–86 of Ecology and Evolution, State University of New York at Juanes F, Smith LD (1995) The ecological consequences of Stony Brook, Stony Brook, NY limb damage and loss in decapod crustaceans: a review Rosenberg MS (2002) Fiddler crab claw shape variation: a geo- and prospectus. J Exp Mar Biol Ecol 193:197–223 metric morphometric analysis across the genus Uca (Crus- Karplus I (2005) Social control of growth in Macrobrachium tacea: Brachyura: Ocypodidae). Biol J Linn Soc 75:147–162 rosenbergii (De Man): a review and prospects for future Sheets HD (2006) IMP: Manovaboard 6.4. Department of research. Aquac Res 36:238–254 Physics, Canisius College, Buffalo, NY Klingenberg CP, McIntyre GS (1998) Geometric morphomet- Simmons LW, Tomkins JL (1996) Sexual selection and the rics of developmental instability: analysing patterns of allometry of earwig forceps. Evol Ecol 10:97–104 fluctuating asymmetry with procrustes methods. Evolution Siva-Jothy MT (1987) Mate securing tactics and the cost of 52:1363–1375 fighting in the Japanese horned beetle, Allomyrina Klingenberg CP, Barluenga M, Meyer A (2002) Shape analy- dichotoma L. (Scarabaeidae). J Ethol 5:165–172 sis of symmetric structures: quantifying variation among Sneddon LU, Huntingford FA, Taylor AC (1997) Weapon size individuals and asymmetry. Evolution 56:1909–1920 versus body size as a predictor of winning in fights be- Knell RJ, Fortey RA (2005) Trilobite spines and beetle horns: tween shore crabs, Carcinus maenas (L.). Behav Ecol sexual selection in the Paleozoic? Biol Lett 1:196–199 Sociobiol 41:237–242 Kodric-Brown A, Brown JH (1984) Truth in advertising: the Sneddon LU, Huntingford FA, Taylor AC, Orr JF (2000) kinds of traits favored by sexual selection. Am Nat 124: Weapon strength and competitive success in the fights of 309–323 shore crabs (Carcinus maenas). J Zool 250:397–403 Lee SY (1995) Cheliped size and structure: the evolution of a Wilson EB (1903) Notes on the reversal of asymmetry in the multifunctional decapod organ. J Exp Mar Biol Ecol 193: regeneration of the chelae in Alpheus heterochelis. Biol 161–176 Bull 4:197–210 Levinton JS, Judge ML, Kurdziel JP (1995) Functional differ- Young RE, Pearce J, Govind CK (1994) Establishment and ences between the major and the minor claws of fiddler maintenance of claw bilateral asymmetry in snapping crabs (Uca, family Ocypodidae, order Decapoda, subphy- shrimps. J Exp Zool 269:319–326 lum Crustacea): a result of selection or developmental Zahavi A (1975) Mate selection — a selection for a handicap. constraint? J Exp Mar Biol Ecol 193:147–160 J Theor Biol 53:205–214 Mariappan P, Balasundaram C, Schmitz B (2000) Decapod Zainal KAY (1990) Aspects of the biology of the squat lobster, chelipeds: an overview. J Biosci 25:301–313 Munida rugosa (Fabricius, 1775). PhD thesis, University of Petrie M (1988) Intraspecific variation in structures that dis- Glasgow play competitive ability: large animals invest relatively Zelditch ML, Swiderski DL, Sheets HD, Fink WL (2004) Geo- more. Anim Behav 36:1174–1179 metric morphometrics for biologists: a primer. Elsevier Petrie M (1992) Are all secondary sexual display structures Academic Press, New York

Editorial responsibility: Victor Meyer-Rochow, Submitted: April 24, 2009; Accepted: February 10, 2010 Bremen, Germany Proofs received from author(s): February 27, 2010